util: fix TextEncoder.encodeInto underfilling

[XadillaX]

TextEncoder.encodeInto() could underfill narrow destination buffers on the optimized path because:

• UTF-16 code units below U+0800 were treated as three-byte UTF-8 sequences starting at U+0400.
• Latin-1 input was exposed as signed char while calculating the scalar tail.
• Surrogate pairs in the scalar tail could be split at the destination boundary.

This change keeps the existing ASCII fast path and replaces the separate prefix-sizing and conversion passes with local bounded conversion helpers. They return both consumed input code units and written UTF-8 bytes, use simdutf for prefixes that are guaranteed to fit, and use scalar conversion at the destination boundary. Malformed UTF-16 is replaced with U+FFFD without creating a temporary well-formed copy.

The helpers are adapted from the APIs added in simdutf/simdutf#1035. They can be removed once Node's vendored simdutf provides those APIs.

Tests cover all destination capacities for Latin-1 and mixed UTF-16 input, including UTF-8 width boundaries, valid surrogate pairs, unpaired surrogates, and exact-fit SIMD chunks. No documentation changes are required.

Benchmark command:

out/Release/node benchmark/util/text-encoder.js \
  op=encodeInto n=1000000 len=256 type=<type>

Input	Before	After
ASCII	30.15M ops/s	30.93M ops/s
Latin-1	14.71M ops/s	19.22M ops/s
UTF-16	9.61M ops/s	10.54M ops/s

Fixes: #65994

[codecov]

Codecov Report

❌ Patch coverage is 96.00000% with 4 lines in your changes missing coverage. Please review.
✅ Project coverage is 90.29%. Comparing base (565f69f) to head (94c65b8).
⚠️ Report is 77 commits behind head on main.

Files with missing lines	Patch %	Lines
src/encoding_binding.cc	96.00%	2 Missing and 2 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #65997      +/-   ##
==========================================
+ Coverage   89.99%   90.29%   +0.30%     
==========================================
  Files         784      789       +5     
  Lines      268410   271453    +3043     
  Branches    51124    51804     +680     
==========================================
+ Hits       241562   245116    +3554     
+ Misses      17385    16825     -560     
- Partials     9463     9512      +49     

Files with missing lines	Coverage Δ
src/encoding_binding.cc	89.80% <96.00%> (+28.69%)	⬆️

... and 119 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[XadillaX]

@anonrig Could you review this when you have time? It now uses local bounded conversion helpers adapted from simdutf/simdutf#1035

[XadillaX]

@jasnell @RafaelGSS Could either of you review this when you have time? This updates the encodeInto path introduced in #60843

[Copilot]

Copilot review overview

🔵 Needs a closer look

The optimized SIMD/scalar boundary logic is complex, and a misleading comment requires correction.

Review effort: Lite
Findings: None

What changed in this PR

Fixes TextEncoder.encodeInto() underfilling in optimized conversions.

Changes:

• Adds bounded SIMD/scalar encoding helpers.
• Handles UTF-8 boundaries and malformed surrogates safely.
• Adds comprehensive destination-capacity regression tests.

File	Summary
test/​parallel/​test-whatwg-encoding-textencoder-encodeinto.js	Adds regression and boundary coverage.
src/​encoding_binding.cc	Reworks bounded encoding and surrogate handling; update the misleading comment about the previous path.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[XadillaX]

@legendecas @aduh95 Could either of you review this when you have time? It updates the TextEncoder encodeInto path, and the CI is green.